devised an alternative route.”
A further advantage of the
Fetch robots is their head for
heights. Safety regulations
in the US require humans
to be harnessed to vertical
conveyance but strapping on
and off devices that resemble
parachutes is time-consuming.
But the Fetch robot can join
the human worker on the
vertical conveyance. The
human hands over objects
and the robot fi ts them
on to the shelves. “This
type of collaborative robot,
called a co-bot, is a recent
development in robotics.
Because they work so closely
with humans, you can see how
workers become fond of them
and start calling them names,”
Smith says.
The future
distribution centres
will all be techbased
Gavin Williams, Managing Director, XPO
Logistics
once carried out by teenagers
looking for part-time work in
the school holidays. Stores
would close once a month and
the manual counting would
begin. “But manual counting
is surprisingly ineffi cient. It’s
about 70% accurate according
to research, whereas the
robots navigate autonomously
around the stores, doing the
data surveys in real time - and
it’s totally accurate,” Smith
says.
The robots need to recharge
their batteries from time to
time, but they work 24 hours
a day and don’t take days off
sick, or get bored, or stop to
chat – or even claim a monthly
wage. “Data survey robots are
a godsend because it’s hard to
fi nd these types of workers,
especially in the US, where
no-one wants to walk up
and down aisles all day long,
measuring stuff,” he declares.
Fetch offers various models
of material handling robots.
The CartConnect models pick
up trollies that deliver heavy
cargo to destinations in the
warehouse. Employees load
the carts, then summon the
robots, who take over the job.
Once the CartConnect robot
arrives at its destination, it
issues a warning signal and
another human operator takes
over in order to complete
the shipping process. “Our
research suggests it can save
more than 20 minutes an hour
per worker and avoids a lot of
the heavy lifting that can be
physically gruelling,” notes
Smith.
Flexibility a key
The more fl exible Fetch
HMIShelf robot has carts
attached that take up to a
dozen small parcels and can
Testbeds, trials and tracking
Advanced robotics is at the
heart of the large warehouse
that XPO Logistics is currently
building for Nestlé, the
world’s largest food and drink
company, near both East
Midlands airport and the
M1 motorway. The 638,000
square foot distribution centre,
in Leicestershire, is billed
as the digital warehouse of
the future. When it opens
next year it will be a testbed
for the development of new
technology and software.
Scientists in a nearby XPO
technology laboratory will
assess the usefulness of more
than 20 smart warehouse
initiatives, including virtual
reality systems. “Those 20
projects will soon become 30,
as we’re developing a new
culture and mindset. We’re
investing US$500m in new
technology at 1,500 sites
globally, but Nestlé will be
the most state-of-the-art yet,”
explains Gavin Williams, the
XPO Logistics’ UK supply
chain Managing Director.
At the heart of the new
warehouse will be the AMR
robots that XPO is developing
with robotics company
GreyOrange. Last year,
XPO announced an initial
US$450m agreement to deploy
the co-bots at sites in the US
and Europe. Williams says
that the GreyOrange AMRs
are both agile and mobile.
They are able to track humans
around the warehouse and
move heavy racks weighing
up to 1,500 kilogrammes.
“They take away the repetitive,
heavy lifting, bring stock to
the humans and allow them
to fulfi ll up to 48 orders
simultaneously,” Williams
says. “These are much more
Robotics isn’t the face of the future. Today’s
applications are ideal for repetitive tasks,
as well as shifting packages
be adapted to pick up larger
boxes. Human operators
load the AMR then send it
on its way using touchscreen
options. Smith reckons that
the robots can be up and
running in hours and are
simple to operate remotely.
“They’re connected to the
cloud so one robot can create a
map of the facility by moving
around, then share it with all
the others. It’s like a puppy
when you throw a ball. It
will fi nd a way to get where
it wants to be and go around
any obstacles. The maps can
be annotated so the pick-up
points and drop-off points are
also shared between robots,
and exclusion zones can be
created,” says Smith.
The next stage is to connect
the robots through the cloud
to software, such as SAP, that
uses machine learning to
draw conclusions about their
performance. One of Fetch’s
customers benefi ted from data
analysis showing the robots
returned to base station much
more quickly than humans
after delivering a load. “The
analysis showed that was the
time when the humans were
stopping to chat on their way
back,” Smith says.
“Another company saw
from the data analysis that
some areas of the warehouse
were congested at certain
times of the day, so they
www.airlogisticsinternational.com June 2019 37
/www.airlogisticsinternational.com
/www.airlogisticsinternational.com